CAMAC Progress Report

                     Week Ending: August 02, 2002
                               Week: 79


Tom Folkers:

Tom Sargent:
     1. Replaced the 19" monitor on the smtocont with a 17" flat 
        panel. It looks better, uses less desk space and consumes half 
        the power of a TV monitor.  The CRTs were also having problems 
        operating at 10,500 ft. altitude.
     2. Added some status variables to the shared memory so that the 
        servo control operator interface could display more useful 
        status information.  I altered the "pause" and "unpause" 
        commands work a little better with the servo control program. 
     3. We observed a tendency for the servo tracking algorithm to 
        sometimes converge very slowly once it gets to within a few 
        arcseconds of the target.  I suspect it may be due to the slew 
        algorithm leaving the telescope closer to the target when it 
        finishes.
     4. There is a run away condition which occurs fairly rarely 
        during slewing - less than once/day.  Print out indicates that 
        it is due to some floating point "nan" results appearing in 
        the slew servo error term.  I can't yet tell what is making 
        this happen.  It appears it is a computational problem in the 
        sun-avoidance code.  I put checks in the software to detect 
        this condition and print out lots of internal information and 
        then stop.  Since then, it hasn't happened again. 
     5. The "suspend" command was written, but not tested. Its purpose 
        is to stop the telescope by making it decelerate to 0 and then 
        having it "track" a stationary object at that point.  The 
        telescope remains ready to move to any other object, but has 
        stopped moving.
     6. A new problem was discovered in the start-up code that cycles 
        the logic on the Krupp rack into a state where the telescope 
        can be driven. It can get hung in one state. I made some 
        changes to fix it, but I can't be sure it is fixed yet.

Pete Hubbard:
     1. SMT to install course encoder filters. Also added .1uF to 
        Krupp rack meter terminals.
     2. Helped Bill with SAMBUS latch option by adding jumpers on pc 
        boards.
     3. Also with looking at oscillation with some SAMBUS signals.

David Forbes:
     1. Last week's digital data glitch has been fixed. I created a 
        buffer enable signal for the parallel port data buffer, got it 
        working right, and now the data bus is quiet. I'm getting 
        clean data from the AOS now.
     2. I updated all four A/D boards with the latest mods and 
        verified their operation on the AOS. I ran the AOS itself from 
        a signal generator to verify the system behavior. It works 
        well.
     3. I checked the data transfer rate with all three A/D cards 
        installed. The parallel port  transfer is busy for 1/3 of the 
        time, leaving 67% processing time.
     4. Today I fed a blanking signal to the AOS controller, and 
        observed that it's not quite working right. I need to work on 
        the blanking logic to fix that. I'll study that code  over the 
        weekend to see what I can do to improve it.